; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
-; RUN: opt -slp-vectorizer -slp-vectorize-hor -slp-vectorize-hor-store -S < %s -mtriple=x86_64-apple-macosx -mcpu=corei7-avx | FileCheck %s
+; RUN: opt -slp-vectorizer -slp-vectorize-hor -slp-vectorize-hor-store -S < %s -mtriple=x86_64-apple-macosx -mcpu=corei7-avx -mattr=+avx | FileCheck %s
;void Distance(float *p1, int p2, unsigned long p3[], float p4[]) {
; long a = p3[0] = 5;
store float %add15, float* %p4, align 4
ret void
}
+
+define void @store_reverse(i64* %p3) {
+; CHECK-LABEL: @store_reverse(
+; CHECK-NEXT: entry:
+; CHECK-NEXT: [[TMP0:%.*]] = load i64, i64* [[P3:%.*]], align 8
+; CHECK-NEXT: [[ARRAYIDX1:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 8
+; CHECK-NEXT: [[TMP1:%.*]] = load i64, i64* [[ARRAYIDX1]], align 8
+; CHECK-NEXT: [[SHL:%.*]] = shl i64 [[TMP0]], [[TMP1]]
+; CHECK-NEXT: [[ARRAYIDX2:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 7
+; CHECK-NEXT: store i64 [[SHL]], i64* [[ARRAYIDX2]], align 8
+; CHECK-NEXT: [[ARRAYIDX3:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 1
+; CHECK-NEXT: [[TMP2:%.*]] = load i64, i64* [[ARRAYIDX3]], align 8
+; CHECK-NEXT: [[ARRAYIDX4:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 9
+; CHECK-NEXT: [[TMP3:%.*]] = load i64, i64* [[ARRAYIDX4]], align 8
+; CHECK-NEXT: [[SHL5:%.*]] = shl i64 [[TMP2]], [[TMP3]]
+; CHECK-NEXT: [[ARRAYIDX6:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 6
+; CHECK-NEXT: store i64 [[SHL5]], i64* [[ARRAYIDX6]], align 8
+; CHECK-NEXT: [[ARRAYIDX7:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 2
+; CHECK-NEXT: [[TMP4:%.*]] = load i64, i64* [[ARRAYIDX7]], align 8
+; CHECK-NEXT: [[ARRAYIDX8:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 10
+; CHECK-NEXT: [[TMP5:%.*]] = load i64, i64* [[ARRAYIDX8]], align 8
+; CHECK-NEXT: [[SHL9:%.*]] = shl i64 [[TMP4]], [[TMP5]]
+; CHECK-NEXT: [[ARRAYIDX10:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 5
+; CHECK-NEXT: store i64 [[SHL9]], i64* [[ARRAYIDX10]], align 8
+; CHECK-NEXT: [[ARRAYIDX11:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 3
+; CHECK-NEXT: [[TMP6:%.*]] = load i64, i64* [[ARRAYIDX11]], align 8
+; CHECK-NEXT: [[ARRAYIDX12:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 11
+; CHECK-NEXT: [[TMP7:%.*]] = load i64, i64* [[ARRAYIDX12]], align 8
+; CHECK-NEXT: [[SHL13:%.*]] = shl i64 [[TMP6]], [[TMP7]]
+; CHECK-NEXT: [[ARRAYIDX14:%.*]] = getelementptr inbounds i64, i64* [[P3]], i64 4
+; CHECK-NEXT: store i64 [[SHL13]], i64* [[ARRAYIDX14]], align 8
+; CHECK-NEXT: ret void
+;
+entry:
+ %0 = load i64, i64* %p3, align 8
+ %arrayidx1 = getelementptr inbounds i64, i64* %p3, i64 8
+ %1 = load i64, i64* %arrayidx1, align 8
+ %shl = shl i64 %0, %1
+ %arrayidx2 = getelementptr inbounds i64, i64* %p3, i64 7
+ store i64 %shl, i64* %arrayidx2, align 8
+ %arrayidx3 = getelementptr inbounds i64, i64* %p3, i64 1
+ %2 = load i64, i64* %arrayidx3, align 8
+ %arrayidx4 = getelementptr inbounds i64, i64* %p3, i64 9
+ %3 = load i64, i64* %arrayidx4, align 8
+ %shl5 = shl i64 %2, %3
+ %arrayidx6 = getelementptr inbounds i64, i64* %p3, i64 6
+ store i64 %shl5, i64* %arrayidx6, align 8
+ %arrayidx7 = getelementptr inbounds i64, i64* %p3, i64 2
+ %4 = load i64, i64* %arrayidx7, align 8
+ %arrayidx8 = getelementptr inbounds i64, i64* %p3, i64 10
+ %5 = load i64, i64* %arrayidx8, align 8
+ %shl9 = shl i64 %4, %5
+ %arrayidx10 = getelementptr inbounds i64, i64* %p3, i64 5
+ store i64 %shl9, i64* %arrayidx10, align 8
+ %arrayidx11 = getelementptr inbounds i64, i64* %p3, i64 3
+ %6 = load i64, i64* %arrayidx11, align 8
+ %arrayidx12 = getelementptr inbounds i64, i64* %p3, i64 11
+ %7 = load i64, i64* %arrayidx12, align 8
+ %shl13 = shl i64 %6, %7
+ %arrayidx14 = getelementptr inbounds i64, i64* %p3, i64 4
+ store i64 %shl13, i64* %arrayidx14, align 8
+ ret void
+}